A common practice in the US is to increase the leakage (or creepage) distance of arresters that are applied in coastal applications.  This is because the salt build up on sheds of arresters can lead to flashover at 60 hz without any surge event.  This flashover can not be mitigated by the arrester and can only be improved on by using longer creepage distance on the arrester.

An alternative to longer creapage distance on the arrester is to put a gap in series with the arrester as shown in this photo.  The polymer housed arrester has a gap and insulator in parallel and both are in series with the arrester.  This is by international (IEC) standards an externally gapped line arrester  (EGLA).  And we thought they were not used in the US.    This scheme effectively reduces arrester housing flashover in the salty environment by removing the voltage from the arrester top and adding the creep distance of the insulator to the arrester creep distance.

Earlier this week ArresterWorks visited a site in SW Miami and photographed a new installation of “Non-gapped Line Arresters” (NGLA) on an FPL line.    FPL has chosen to improve this line with the installation of the NGLA only on the top phase.  The installation looks very well done.

Cross Arm with Fire Damage How did it happen?

Occassionally I come across a pole fire that is associated with an arrester.   The main question in the above photo is what caused the fire damage.   Was it the arrester or insulator.   And how do you suppose the ground wire got separated by so much?

Would love to hear you opinion.   Email me at jonathan.woodworth@arresterworks.com

It is well known that the fundamental purpose of an arrester is to limit the voltage across insulation it is electrically connected in parallel with.  It is however not as well known how to verify this function.  There is no easy way to demonstrate the function of a high current surge arrester typically found on power systems so surge generators are employed.  To demonstrate the clamping effect at high currents, the Marx Generator is used. In its simplest form it is a charged capacitor of substantial capacity is rapidly discharged by a gap type switch through the arrester. The wave shape of the discharge is controlled by series resistance and series inductance.   The voltage across and the current through the arrester are monitored and used as verification of the arrester clamping function.   A typical discharge looks like the following:

In 2009 I had the opportunity to visit Hermsdorf Germany. While there I learned that this city was the place where Erwin Otto Marx invented his now famous Marx Generator. Anyone who has ever impulsed an arrester will probably recognize the name. The folks in Hermsdorf are quite proud of the fact that Marx invented the generator there and they should be. Their present city hall is in the building where Marx performed his experiments on the generator. As you can see from the photo below they even have an electrical insulator mounted on the building

Comment  from Professor Hinrichsen (TU Darmstadt)

Marx became head of the R&D laboratories of HESCHO, located in Hermsdorf, in 1923. There he invented the impulse generator, today referred to as “Marx generator”. In 1925, Marx was appointed a professorship at Braunschweig (Brunswick) Technical University, where he worked until 1945, and after a short interruption, until 1962. His main research before and during World War II was on HVDC transmission. He was working on pressurized gas arc rectifiers, which were never realized commercially, because mercury valves turned out to be much better.

For those of you interested in Modeling this generator in ATP, below is the schematic and  here is a link to the actual model.

Here you will find a photo of two marx generators at the Gerogia Tech NEETRAC Labs

Glass Housed Arrester

After more than 50 years this glass housed arrester is still protecting its transformer

The glassed housed silicon carbide arrester pictured above was introduced in the late 1930’s by the Line Material Company and produced from then until 1958.  The factory where it was last produced was and still is located in Olean NY and is now part of Cooper Industries.  This arrester had a significant influence on the  factory location because the two most significant raw material suppliers were also in Western New York which in the 1950’s was a more important consideration than it is today.  The glass housing was produced by the Corning Glass Company then and still located in Corning NY, 80 miles to the east, and the silicon carbide material (black granular material in the bottom half of this unit ) was produced by the Carborumdum Company in Niagara Falls NY, 80 miles to the north.  This arrester model design was modified in minor ways over its production life of 20 years, but the glass housing was a main theme the entire time.  To this authors knowledge, the Line Material Company was the only arrester manufacturer to use a glass housing.  The arrester was designed in Milwaukee Wisconsin by Ralph Earl and Alwin G Steinmayer both of which were granted patent  2,165,964 for the design in 1936.   In this photo  it stands more than a half a century later still guarding the transformer  that is most likely in the same age category.  The glass housed arresters produced in the 40’s and 50’s can still be found around the world.  It is estimated that more than a million units were put in service during the production years.

The rationale for using a glass housing was simply to make the internal parts visible so if the arrester was failed or  damaged, it could be easily observed without testing.  On the top half if this unit, the  shiny brass electrodes that make up the gap assembly clearly indicate that this arrester is in pristine condition.  Clearly (no pun intended) this unit is still in good working condition.  The transformer was likely challenged by lightning numerous times during its life and it looks like its Sentry did its job since it is still powered and loaded.

Patent 2165964 (1936)

I find it amazing how well arresters withstand the ice build up during the winter months.  Here are a couple of photos to enjoy.

How many arresters are in this photo and what type are they?????

And the Answer Is……

.

.

This ArresterFacts can be downloaded in PDF here

.

Figure 1 Cutaway of Deadfront Arrester with detailed component functions

The Deadfront (Figure 1)  and Separable arresters are one of the latest types of arresters to be introduced to the surge protection world.  It was not until the introduction of the MOV type arrester that this type of  arrester as it is known today was possible.  In 1979 a US patent was issued to Francis Cunningham and appears to be one of the earliest description of this device.

Figure 2 Early Deadfront Arrester Patent Sketch - Cunningham 1979

This ArresterFacts covers the fundamentals of these devices as well as the reasons why they are such important devices in the protection of cable systems.  There are several other names for these devices including elbow arresters, parking stand arresters, and bushing arresters in the IEEE market. In the IEC market, the term separable arrester is the only term used to describe this type of arrester.   In all cases, they are used to protect cable systems in the medium voltage range 11-42kV systems.

Definitions

Deadfront Arrester: An arrester assembled in a shielded housing providing system insulation and conductive ground shield, intended to be installed in an enclosure for the protection of underground and pad-mounted distribution equipment and circuits.  Typically 200A load break versions can be connected or disconnected with the system energized however 200A deadbreak designs do exist.  (IEEE)

Separable Arrester: An arrester assembled in an insulated or screened housing providing system insulation, intended to be installed in an enclosure for the protection of distribution equipment and systems. Electrical connection may be made by sliding contact or by bolted devices; however, all separable arresters are dead-break arresters.    (IEC)

The Voltage Doubling Effect

The deadfront  and separable  arrester are special arresters used primarily to mitigate the real and ever present voltage doubling effect at open points on underground circuits.

When a lightning surge hits a riser pole arrester, the arrester clamps the surge, but allows an ongoing surge with the magnitude of the clamping voltage of the riser pole arrester to travel on and into the underground circuit.

Figure 3 shows a circuit that has two underground branches in a distribution underground circuit.  Branch 1 has a deadfront arrester at its end point and Branch 2 has no endpoint arrester.

Figure 3 Underground circuit with Deadfront Arrester at Open of one branch and not on other branch

When the ongoing surge from the riser pole arrester meets an endpoint in the underground circuit, it will double in magnitude at that end  point in the circuit as shown in Figure 4.

Figure 4 Voltages along unprotected Branch showing Voltage doubling at endpoint without deadfront arrester

When the same surge meets the endpoint in Branch 1 where a deadfront arrester was installed, the voltage is not doubled and is in fact decreased as shown in Figure 5.  Even though the riser pole arrester and deadfront arrester are of the same rating, the clamping voltage at the deadfront is much less than at the riser pole because the riser pole arrester conducted nearly the full lightning stroke current as shown in Figure 6 and the deadfront arrester conducted only a few kA.

Figure 5 Voltages along branch 1 with deadfront arrester

Figure 6 Current through the Riserpole arrester and Deadfront arrester during a 100kA lightning strike

Deadfront Arrester Not at End Point

In the case where a deadfront arrester is not located  at the endpoint as shown in Figure 7 the results are interesting.  The end point voltage with no arrester still doubles but is only double the level of the deadfront arrester as displayed in Figure 8 and not double the level of the riser pole arrester as shown in Figure 4.

Figure 7 Circuit with Deadfront arrester one cable span back from end point

Figure 8 Voltage along Branch 1 with deadfront arrester back one span of cable

.

Arrester Rating Considerations

The Deadfront and Separable arresters are often rated as  5kA arresters.  The IEEE Deadfront arrester may also be rated as a light duty arrester which is one energy handling class below the normal duty arrester ( 5kA arrester).  These lower energy ratings are quite appropriate since underground circuits are universally protected with an arrester at the riser pole and as seen in Figure 6 the riser pole arrester conducts the vase majority of the surge current, leaving only a few kA of current for the deadfront arrester.

A unique arrester characteristic of separable arresters and deadfront arresters is the configuration of the high voltage terminal of the unit.  Since the high voltage end must mate intimately with a bushing, the designs are very specific to the available bushings in both the IEEE and IEC markets.   In the IEEE market, there are three common interfaces, 15, 25 and 35kV.   In the IEC market, several interfaces also exist. Among others, they include the 250A, 400A, and 630A, all are deadbreak.

This type of arrester is also configured in several formats depending on how it must be installed in the circuit.  For example the parking stand style is a design that is mounted in a bracket known as a parking stand of a pad mounted transformer and can be seen in Figure 9 above .  An incoming line can then easily be terminated directly into the arrester thus avoiding an open endpoint. Other variations include inline arresters where the incoming line feeds through the arrester and then into the transformer.  In all cases, the arresters perform identically.

Figure 9 Two common mounting configurations

Failure Mode Considerations

The failure mode of this arrester is tested differently than all other arrester types.  For this arrester, it is acceptable for the arrester to eject disks out the bottom of the unit but not out the side during a failure event.  This is only acceptable because the arrester is generally installed in enclosed areas where ejection of parts would not be a safety issue as it would be for an overhead arrester.  The fault currents used for testing is also much lower since it is uncommon for very high fault currents to be available in underground distribution circuits.

Installation Considerations

As with all arresters it is important for the ground lead of the arrester to be as short as possible.  It should also be attached in such a way that it is not stressing the arrester.

Relevant Test Standards and Application Guides

The three IEEE standards that apply to this arrester are C62.11, IEEE 592 and IEEE 386 which describe all the necessary tests that are required for certifications. Application Guides C62.22 and C62.22.1 are the appropriate application guides.

The IEC documents of relevance are 60099-4 and CENELEC 629.1 S1 for testing and 60099-5 for applications.

***************************************

ArresterFacts Usage

ArresterFacts are Copyrighted documents intended for the education of arrester users and stakeholders.   If you choose to copy any part of this document for teaching purposes you have my permission, however please give ArresterWorks proper credit.

Thank you for using www.ArresterWorks.com as your source of information on high voltage surge arresters.

Jonathan Woodworth

Principal Consultant

ArresterWorks

A sheath voltage limiter is an arrester that is used to clamp the voltage induced on the sheath of an underground high or medium voltage cable during a fault on the system.  It is common practice when applying underground cable to only ground the cable sheath at one end of the cable, and leave the other end open.  Leaving one ungrounded reduces circulating currents and losses during steady state operations.  IEEE standard 575-1988 indicates that typically distribution arresters can be used for this function.  IEC Application guide 60099-5 soon to be published offers a formula to use to determine the voltage rating of the arrester, but does not offer any guidance regarding the energy handling rating.

The photo below is an SVL on a 69kV underground system in Lakeland Florida.  A standard distribution arrester is being used.  This is the first ever SVL I have seen on a transmission line riser pole.

Finally in November of last year I was able to get ATP loaded on my windows 7 device.  That was no small task, in fact I required the help of a more computer proficient person to do it.   In the end, he had to change some  global variable in the Window 7 system.     Way out of my computer scope of understanding.     BUT, what a surprise ATP has been.  I have been working with it nonstop since then.   The big surprise is how effective it is at teaching about arresters and surges on power systems. Using ATP should be a required skill for any power engineer especially if they want to understand transients on the systems.

For those who are not familiar with ATP, it is an open-source program used to model and simulate power system operation.  It is maintained by a consortium of professors around the world. The only requirement in using it is that you do not sell it.  Once you retain a license and agree to follow their no sell rule, you get the passwords to the databases where you download the program and other helpful documents.

The program version that I am using is ATPDraw and it is a graphic interface to the ATP program that does all the analysis.  I have been graphing the output using XYPlot.  Once you learn the basics of how to use these programs and how they interact, you can build circuits with any type of power system product you want.  Hit the run button and out comes the response of the system.

Yesterday I ran a test on a 10 pole transmission line to examine the current sharing  of arresters when there is a shield failure and a direct strike to the phase conductor.  I am amazed how much sharing goes on.   The arrester models are very touchy, so I have been using the nonlinear resistor component for arresters.  As long as I use the same VI characteristics, they seem to work well.

Anyone with arrester modeling experience could chip in here and offer their experience.

I highly recommend it for anyone that wants to understand transients and their effect on power systems better.

go to www.empt.org for more information on how  to download and have fun.

Comments please.

The first time you comment, it will not show up immediately, but once you are an approved commenter, then you can comment as much as you wish..